- Reliability
- 70.7%
- Response time
- 1.66s
- Cost / min
- $0.0798
Azure Realtime is ahead on success rate, calls where the agent went quiet, interruption and response time, and GPT-Live 1 + GPT-6 Sol on reliability, agent responses, data accuracy and cost.
Last updated Methodology by Dileep Chagam
Our guide covers testing OpenAI Realtime API voice agents.
Scenarios the agent passed on every one of its 3 runs.
GPT-Live 1 + GPT-6 Sol is more reliable, by 4.9 points. Azure Realtime passed 58 of 82 scenarios on all three runs, and GPT-Live 1 + GPT-6 Sol passed 62. A scenario only counts when every run of it passed.
GPT-Live 1 + GPT-6 Sol is more reliable, by 4.9 points. Azure Realtime passed 58 of 82 scenarios on all three runs, and GPT-Live 1 + GPT-6 Sol passed 62. A scenario only counts when every run of it passed.
For a voice agent, GPT-Live 1 + GPT-6 Sol, since it gets more scenarios right on every run. Azure Realtime is ahead on success rate, calls where the agent went quiet, interruption and response time, and GPT-Live 1 + GPT-6 Sol on reliability, agent responses, data accuracy and cost.
GPT-Live 1 + GPT-6 Sol, by 3.0 points. Azure Realtime saved everything exactly right on 86.1% of the calls that save data, and GPT-Live 1 + GPT-6 Sol on 89.0%. On Booking calls, 94.2% against 100.0%. On Medicare calls, 65.2% against 60.6%. A call counts only when nothing is missing, wrong, extra or saved twice.
Yes. Azure Realtime starts replying 0.36s sooner at the median. Azure Realtime takes 1.66s at the median and 2.22s at p90; GPT-Live 1 + GPT-6 Sol takes 2.02s and 2.42s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.
Azure Realtime, by 0.08 on a 0 to 5 scale. Azure Realtime scores 4.96 and GPT-Live 1 + GPT-6 Sol 4.89 for how well the agent handled being talked over.
Azure Realtime, by 8.1 points. The agent went quiet for 10 seconds or more on 4.1% of Azure Realtime calls and 12.2% of GPT-Live 1 + GPT-6 Sol calls.
GPT-Live 1 + GPT-6 Sol is $0.0245 cheaper a minute. Azure Realtime costs $0.0798 a minute of call and GPT-Live 1 + GPT-6 Sol $0.0553, at each vendor's published rates.
Cekura ran both as the whole agent in the same open-source Pipecat pipeline, with the same prompt, tools and connection. A simulated caller worked through all 82 scenarios on live calls, three times each, 3,690 calls across the benchmark. A call passes only when the agent said the right things and saved the right data.
Cekura runs these simulated calls on your agent and scores every one.